Fig. 3 depicts the group mean percentage error rates committed by humans in the congruent (good) and non-congruent (bad) phrasing groups over acquisition (top panel) and divided by response type (pattern AT-101 responses, phrasing cue response) over their 10 respective training patterns (bottom panel). Fig. 4 shows the group mean percentage of errors committed by humans on each pattern element for their respective patterns. As evidenced by the figures and the data analyses described below, both groups learned to perform their patterns at levels better than would be predicted by chance. However, the group learning with congruent phrasing showed facilitation by making fewer errors (11.96%, SEM = 3.7) relative to the group learning with non-congruent phrasing (40.26%, SEM = 3.44), as indicated by mean percentage error rates over pattern element responses. With respect to response type, those learning the congruent pattern committed significantly fewer errors than those learning the non-congruent pattern across the pattern element responses and the phrasing cue response. In examining these errors, significantly more errors were committed on the first element of each chunk and on the phrasing cue. The analysis of errors across the entire pattern by pattern element supports this conclusion.
↧